The scariest thing about data science is that people think it is scary.
Data science is the talk of the town in the present-day world. It came into the picture in 2008 when with the advancement of internet and device connectivity, an immense flow of data was observed. In recent years, data science has been advancing our technological growth at a fantastic rate. But with any new technology, whether it be guns, knives, bows, spears, fire, or even data science, comes malcontent. The scariest thing about data science is that people think it is scary. This article features the ten scary truths about data science.
Knowledge:
Most data scientists and the organizations that employ them don’t seem to understand how data science is actually done, nor what it is exactly. They sort of jumped on the bandwagon — without really understanding it, nor why it was important to them in a very visceral way.
Labels:
Now say, you want to know if a person is going to commit a crime or not in the future. For the people you already know, your label will be whether that person committed a crime or not.
Penalty System:
This penalty is for the machine. While learning from the Data and the Labels, the machine makes errors, and to make sure it learns from its mistakes, you have to come up with a formula to tell what kind of error receives what kind of penalty. Some errors you may allow the machine to commit without much penalty, and some other errors are heavily penalized.
Data science leadership is sorely lacking:
Most executives in charge of data science decision-making are neither educated nor trained in actual data science theory and techniques. Instead, they have relied upon non-data-driven, plug-and-play features that can be launched promptly.
Few teams have a Head of Data, Data Science Manager, or another relevant role. As a Data Scientist, you may report to someone specialized in just product, engineering, or even another discipline.
Here are some of the challenges of data scientists:
Finding the data:
Finding the right data is still the most common challenge for data scientists, directly impacting their ability to build strong models. Do most companies collect tremendous volumes of data without determining whether it is consumable or not? This makes it harder for data users to find the truly relevant data assets for the business strategy. Data is scattered across multiple sources, making it difficult for data scientists to find the right asset. That’s why so many companies use a data warehouse, in which they store the data from various sources.
Getting access to the data:
Security and compliance issues are making it harder for data scientists to access datasets. Like confidential data is becoming vulnerable to cyber-attacks, data scientists struggle to get consent to use the data, which drastically slows down their work, worse when they are refused access to a dataset.
Understanding the data:
When data scientists find and obtain access to a specific table, they can finally work their magic and build powerful predictive models. Undocumented assets roam around your business with unproductive data scientists spending 80% of their time trying to figure them out.
Right communication:
Communication is pivotal to forging a successful career for the data scientist. Working closely with the company’s decision-makers and maintaining a solid relationship is essential. Always look for an opportunity to solve the business problem or in-house team concerns with a chance for automating redundant tasks or basic data retrieval. Most data science professionals in a company, by default, will be considered analytics and data experts.
Data cleaning:
Data scientists spend most of their time pre-processing data to make it consistent before analyzing it, instead of building meaningful models. Because real-life data is nothing like hackathon data or Kaggle data. It is much messier. This task involves cleaning the data, removing outliers, encoding variables, etc. The worst part of a data scientist’s career is data pre-processing, which is crucial because models are built on clean, high-quality data. Otherwise, machine learning models learn the wrong patterns, ultimately leading to wrong predictions.
Communicating with non-technical stakeholders:
Data scientists’ work is meant to be perfectly aligned with business strategy, the goal of data science is to guide and improve decision-making in organizations. Hence, one of their biggest challenges is to communicate their results to business executives. Data scientists often have a technical background, making it difficult for them to translate their data findings into clear business insights.
Source: analyticsinsight.net
